Tag STS in the AZF Region Associated With Azoospermia in a Tunisian Population


Laboratory of Human Molecular Genetics, Faculty of Medicine, Sfax, 3029, Tunisia (e-mail: lobna_h@lycos.com).


ABSTRACT: The aim of this study was to determine the appropriate tag sequence tagged site (STS) associated with azoospermia. We evaluated the incidence of Y chromosome microdeletions in Tunisian infertile male patients by polymerase chain reaction using 14 STSs in the azoospermia factor (AZF) region of Yq11. A logistic regression analysis was performed to test the association of STSs with semen quality. Haploview version 3.11 was used to identify the possible blocks of deletion involving a minimum number of STSs and that can be used to tag the deletion block in future analysis. Using the 14 STSs, 48% infertile patients (102 of 210) had microdeletions of Y chromosome but, following the European Academy of Andrology guidelines, only 16% of patients had microdeletions. A statistically significant difference was found with some STSs for azoospermia and oligozoospermia. A candidate combinaison composed of 4 STSs (RBMY-sy157-sy84-sy130) was associated with azoospermia in a Tunisian population. According to this study, this tag can be used in the screening of Y chromosome microdeletions before assisted reproduction treatment in a Tunisian population.

On average, 10% of couples encounter difficulties procreating and consult infertility clinics. In about half of cases, sperm production is defective, either quantitatively or qualitatively. Approximately one third to one half of these cases are classified as idiopathic (Hargreave, 1994), but they may have an unidentified genetic anomaly. The main function of Y chromosome is the propagation of species through sex determination and control of spermatogenesis. Long ago macroscopic deletions in the long arm of Y chromosome were suggested to be responsible for azoospermia (Tiepolo and Zuffardi, 1976). With the advancement in molecular biology, 3 nonoverlapping regions referred to as “azoospermia factor” (AZFa, AZFb, AZFc from proximal to distal Yq) have been defined as spermatogenesis loci (Vogt et al, 1996). AZF microdeletions are associated with azoospermia, oligozoospermia, and also with a varied testis histologic profile ranging from Sertoli cell only, hypospermatogenesis, to maturation arrest (Vogt et al, 1996).

In the last few years, the Y chromosome—specific sequence tagged site (STS) markers were used in a great number of clinical studies to identify microdeletions in DNA from peripheral blood lymphocytes (Kleiman et al, 2001) and from sperm (Hellani et al, 2005) of infertile males. Many publications have determined the prevalence of such deletions in male infertility. Most of the studies mainly focused on patients with sperm count deficiencies. The average Y microdeletions for infertile males were 8.2%, and most deletions were associated with azoospermia (Foresta et al, 2001). However, a few studies screened a more random group of men (Krausz et al, 1999; Van Landuyt et al, 2000). A few nonpolymorphic deletions were found in individuals with higher sperm counts (5 × 106−20 × 106/mL) (Kent-First et al, 1999), even normozoospermia (more than 20 × 106/mL) (Pryor et al, 1997) and teratospermia (Hellani et al, 2005), and also in a few nonidiopathic infertile men with abnormalities such as undescended testes, obstruction of vas deferens, and varicocele (Simoni et al, 1998). As a result it is unclear whether a wider selection of patients attending a fertility clinic should undergo genetic screening. More studies need to be carried out on patients with a range of phenotypes to clarify the value of routine screening.

The number of STSs used for the detection of Y chromosome microdeletions varied from one study to another: from 8 STSs (Martinez et al, 2000) to 85 STSs (Pryor et al, 1997). Therefore, it is important to define a tag STS instead to use an important number of STSs in the screening of Y chromosome microdeletions before assisted reproduction treatment. The aim of this study was to determine a tag STS associated with azoospermia in a Tunisian population.


Patient Selection

The study population consisted of 210 infertile patients seeking andrologic investigation for couples infertility at the Laboratory of Histology of Faculty of Medicine. They were subject to detailed clinical and biological investigations, including cytogenetic and endocrinology studies, physical examination and, when possible, histology of a testis biopsy. Patients having chromosomal abnormalities and obstructive azoospermia were excluded from the study. On the basis of spermiogram, individuals (n = 210) were subdivided into 3 groups: azoospermic (0 spermatozoa [spz]/mL), oligozoospermic (less than 20 × 106 spz/mL), and normozoospermic (more than 20 × 106 spz/mL) according to the criteria of the World Health Organization (World Health Organization, 1999).

Molecular Investigations

Genomic DNA was extracted from peripheral blood lymphocytes using standard techniques. Polymerase chain reaction (PCR) analysis was performed with 50 ng of genomic DNA, 2.0 mM MgCl2, 1× reaction buffer from Taq DNA polymerase Promega, 200 μM dNTPs, and 1 unit of Taq DNA polymerase. Thermocycling consisted of an initial denaturation of 5 minutes 94°C followed by 35 cycles of 40 seconds at 94°C, 50 seconds at 55°C, 50 seconds at 72°C and, finally, 10 minutes at 72°C. Negative PCR amplifications were repeated at least 3 times to confirm the deletion of a given marker. A healthy fertile male and a female were used as positive and negative controls for each STS amplification. PCR products were analyzed on a 2% agarose gel. The STS primers tested on each subject are shown in Table 1 with their relative position on the physical map.

Table 1. . Order of markers and their relative position on the physical map
LocusSTS No.MarkerPhysical Position*Distance Between Markers (bps)Size (bps)Cytogenetic Band
  1. *Data available from Genome Bioinformatics Group (2003)

AZFa1Sy8113407162 to 1340737113407162209Yq11.21
 2DFFRY13847050 to 13847475439679426Yq11.21
 3SY8613918257 to 1391857470782317Yq11.21
 4DBY114033337 to 14033613114763277Yq11.21
 5SY8414100173 to 1410051866560326Yq11.21
 6SY8714319162 to 14319579218644252Yq11.21
 7SY18215298960 to 15299084979381125Yq11.221
AZFb8SY12721717107 to 217173806418023274Yq11.223
 9SY13022089075 to 22089247371695173Yq11.223
 10SY13422702751 to 22703053613504301Yq11.23
 11RBMY23307689 to 23308407604636299Yq11.223
AZFc12sy25524160002 to 24160125851595124Yq11.223
 13Sy25424161378 to 241617571253380Yq11.223
 14SY15725022148 to 25022437860391289Yq11.223

Statistical Analysis

The χ2 test in 2 × 2 contingency tables for each marker was calculated to determine which STS was associated with azoospermia and oligozoospermia. A logistic regression analysis was also performed to test the association of STSs with semen quality in these possible models: 1) normozoospermic group vs oligozoospermic group and 2) normozoospermic group vs azoospermic group.

Significance of P values was assessed using a Bonferroni correction (tests with P < .05/14 = 0.0035 are considered significant).

SPSS version 10 was used in all statistic analysis.

Haplotype Analysis

Haploview version 3.11 (Barrett et al, 2005), available online at www.hapmap.org, was used to estimate haplotype frequencies and blocks of deleted STSs. Of note, because we are dealing with the male-specific part of the Y chromosome (MSY), the concept of haplotypes is defined as a combination of deleted or undeleted STSs and not as commonly understood because there is no recombination. Besides, the blocks of STSs that are deleted together or undeleted together are determined using the same methods of Haploview used for calculating blocks of linkage disequilibrium. The objective of this analysis was to identify the possible blocks of deletion that may be followed by a minimum number of STSs in this block. These STSs will thus be used to tag the deletion block in future analysis.

Haplotype Diversity

We calculate the STS haplotype diversity by H = 1-Σpi2, where pi is the frequency of haplotype i, and the sum is all the observed haplotypes.


Clinical Findings

Of the 210 patients included in this study, sperm counts revealed azoospermia (n = 53, 29%), oligozoospermia (n = 72, 33%) and normozoospermia (n = 85, 38%).

We found that 102 of 210 (48%) patients had noncytogenetically detectable microdeletions. However, following the European Academy of Andrology (EAA) guidelines we found that only 16% had Y chromosome microdeletions (HadjKacem et al, 2006). The position and extent of the deletions are shown schematically in Figure 1a. Analysis of our results showed that all microdeletions were located within AZFa, AZFb, and AZFc regions. Most of these deletions included separate markers. However, 3 patients had an extended deletion, into AZFa for 2 patients (STSs 1–4 and STSs 2–6 respectively) and into AZFb and AZFc for the third patient (STSs 10–13). These extended deletions were associated with azoospermia and oligozoospemia.

Figure 1.

. A schematic representation of the human Y chromosome showing the (a) position and extent of deletions and (b) number and frequency of deletions haplotypes in all groups. Azoo indicates azoospermia; Oligo, oligozoospermia; Normo, normozoospermia; Nber: number; Freq, frequency; H1, haplotype; *, polymorphic markers; underlines, haplotypes present only in azoospermia, oligospermia, and normozoospermia groups; bold, haplotypes more frequent in azoospermia, oligospermia, and normozoospermia; □, deleted; ▄, not deleted; -, not tested.

Relation: Male Infertility—STS Microdeletions

Using the logistic regression, we found that the deletion of specific STSs was associated with azoospermia or oligospermia with a probability of less than .5 (Table 2). Indeed, azoospermia was found associated with the deletions of STS 1, STS 2 (P = .001 for both STSs), and STS 11 (P = .0001). Oligozoospermia was found associated with the deletions of STS 1, STS 2, STS 11, STS 13, STS 14 (P = .0001 for all markers), and STS 4 (P = .001).

Table 2. . Results of χ2 test for all STSs*
STS No.χ2Pχ2Pχ2Pχ2P
  1. *Bold indicates significant P value based on a Bonferroni correction (P < .05/14 = 0.0035)


Combinations of STS Deletions Associated With Azoospermia

The different combinations of microdeleted STSs were analyzed in the 3 groups of patients. Each combination was considered as marker haplotypes, and the term haplotype is subsequently used. Figure 1b shows all the 58 haplotypes (H1 to H58) observed in the sample of 210 individuals. A total of 13, 29, and 2 combinations were found only in azoospermia, oligospermia, and normozoospermia groups, respectively. Two combinations (H15, H55) were found to be more frequent in the azoospermia group, 1 combination (H19) in the oligozoospermia goup, and 2 combinations (H45, H46) in normozoospermia group.

To identify a putative tag STS associated with azoospermia, the 2 oligozoospermia and normozoospermia groups were compared with the azoospermia group. Using Haploview, 3 haplotype blocks were obseved in azoospermia goup. Block 1 spans 115 kilobases (kb) of the Y chromosome and is defined by 2 STSs (STS 3 and STS 4). Block 2 spans 613 kb and contains 2 STSs (STS 9 and STS 10). Block 3 spans 853 kb and is defined by 3 STSs (STS 11-STS 13). Besides, 3 blocks were apparent in the oligonormozoospermia group. Block 1 spans 7616 kb and contains 4 STSs (STS 5-STS 8). Block 2 spans 604 kb and is defined by 2 STSs (STS 10 and STS 11). Block 3 spans 862 kb and contains 3 STSs (STS 12-STS 14). Table 3 shows the results of logistic regression with the azoospermia group in comparison with the oligonormozoospermia group. A statistically significant difference was found for the deletion of STS 5 (P = .032), STS 9 (P = .05), STS 11 (P = .004), and STS 14 (P = .011). When we combined the results of Haploview and those of logistic regression, we identified 2 tags for STSs: 3-5-10-11 and 9–14, which are potentially associated with azoospermia.

Table 3. . Results of χ2 test for all STSs by comparison of 2 groups (azoospermia and oligonormozoospermia groups) (P < .05)
STS No.χ2P
  1. *Significant difference between the 2 groups (P < .05)


The haplotype diversity value was calculated adding the STS one by one (Table 4). The observed haplotype diversity value was 77.4% in the azoospermia group and 70.5% in the oligonormozoospermia group. In the azoospermia group, when adding the STS one by one, the number of different haplotypes rose from 2 to 22. However, in the oligonormozoospermia group the number rose from 2 to 46. The addition of STS 10 had no effect on haplotype diversity value: it was only slightly changed (from 40.3% to 41.6% in the azoospermia group and from 25.5% to 29.8% in the oligonormozoospermia group). Similar results were found with the addition of STSs 3, 4, 7, 12, and 13. Consequently, we retained the haplotype of STS 11-14-5-9.

Table 4. . Haplotype number and diversity values in azoospermia and oligonormozoospermia groups by combining the STSs one by one
Haplotype compositionNo. of Observed HaplotypesHaplotype Diversity (%)No. of Observed HaplotypesHaplotype Diversity (%)
  1. *Bold numbers indicate tag STS


The “undeleted” tag for STS 11-14-5-9 was found to be the most frequent in the 2 groups. Besides this tag, 2 tag STSs showed different frequencies in the 2 groups. The tag with a single deletion in STS 9 was twofold more frequent in the oligonormozoospermia group than in the azoospermia group, while the tag with a single deletion for STS 11 showed the inverse pattern, being sevenfold more frequent in the azoospermia group (Figure 2). Notably, the frequency of STS 9 was less in the azoospermia group than in the normooligozoospermia group. However, the deletion of STS 11 was clearly more frequent in azoospermia than in oligonormozoospermia, especially when considered together with the deletion of STS 5. The tag with either or both STSs (11 and 5) deleted had a frequency of 24.6% in azoospermia against 3.1% in oligonormozoospermia group.

Figure 2.

. Frequency of haplotype for STS 5-9-11-14 in azoospermia and olignormozoospermia groups. Bold indicates discordant frequency of haplotypes between the 2 groups.


This study tried to evaluate the incidence of Y chromosome microdeletions in Tunisian infertile male patients and to determine the appropriate tag STS associated with azoospermia.

The reported frequency of deletions varies from 1% (Van der Ven et al, 1997) to 55% (Foresta et al, 1998) mainly depending on inclusion criteria, but most studies reported an incidence below 15%. Using 14 STSs, we found a relatively high frequency of deletions (48%), but using the recommendations of the EAA guidelines (Simoni et al, 1999) we had found an overall deletion frequency of 16% (HadjKacem et al, 2006). Those showing deletions had AZFc (sy254, sy255) in 11%, AZFb (sy127, sy134) in 7.3%, and AZFa (sy84, sy87) in 6.7%. However, in the present study we failed to see any deletions in the sy86 and sy84 STSs, commonly recommended for analysis. On the other hand, the missing STSs were sy87, sy81, and the genes DFFRY and DBY, which are interspread between or around sy86 and sy84. Hence, the chances of finding a deletion would be higher if more sets of primers are used, as suggested by Thangaraj et al (2003) and Mitra et al (2006). However, this contrasts with the recommendations of the EAA guidelines (Simoni et al, 1999), which suggest that more than 90% of microdeletions can be detected with the use of only 2 STS markers for each AZF loci.

Five STS markers were deleted in normozoospermia group: STS 3, 6, 8, 9, and 10. These deletions can be considered as nonfunctional variations because they occurred in all groups. Similar results were found by Hellani et al (2005), who reported a deletion of STS 5, 6, 10, 9, 13, and 14 in the normozoospermia group, with a sperm count of more than 20 × 106 spz/mL. In our study the deletions of all other STSs can be considered functional and important in the development of spermatogenesis, because they were significantly associated with a severe spermatogenic phenotype. Indeed, azoospermia was associated with the deletion of STS 1, 2, and 11, and oligozoospermia was associated with the deletion STS 1, 2, 4, 11, 13, and 14.

To identify a tag STS associated with azoospermia, the 2 groups' oligozoospermia and normozoospermia were compared with the azoospermia group. A significant difference between normooligozoospermia and azoospermia groups was found for the deletion of STS 5, 9, 11, and 14 (sy84, sy130, RBMY, and sy157). The block of tagging STS composed by STSs 11-14-5-9 (RBMY-sy157-sy84-sy130) was retained to look for associated deletion patterns. This block consisted of an STS from each of the 3 regions of AZF and an STS of the RBMY gene. There is no previous finding about the association of a tag STS and azoospermia. We found that combinations having a deletion in STS 5 and STS 11 (separately or together) are clearly more frequent in the azoospermia group than in the oligonormozoospermia group. The combination with both STSs deleted was absent in the oligonormozoospermic group while having a relatively high frequency (5.7%) in the azoospermic group. This may indicate that a certain degree of severity is conferred by each deletion, which is maximal when both are deleted. This makes it a good candidate for further functional studies. The combination with single deletion of STS 9 or STS 11 had an inverse pattern between the azoospermia and oligonormozoospermia groups. Because the deletion of STS 9 was frequent in oligonormozoospermia, it could not be functionally associated with azoospermia, confirming the result of logistic regression that STS 9 was polymorphic. It may be possible that this deletion has a protective effect. Such an explanation is only tentative because it was never before suggested.

Due to the prognostic value of this type of deletion, it can be recommended for a screening of microdeletions in Y chromosome using the tag STS 11-14-5-9 (RBMY-sy157-sy84-sy130) before assisted reproduction treatment in our population. The study of microdeletions for other STSs in the interesting gene RBMY would be very useful to determine if a haplotype potentially associated with azoospermia could be identified.


We are grateful to Dr Ahmed Rebai for his help in statistical analysis, valuable suggestions, and comments on this work. The authors thank Mr Ayed Hajji from the engineering school of Sfax for his help with English.


  1. This work was supported by ministère de la recherche, de la technologie et du développement de compétence (Tunisia).